Smoking habit from the paternal line and grand-child's overweight or obesity status in early childhood: prospective findings from the lifeways cross-generation cohort study.

Abstract

The role of smoking from the paternal line during the pre-conception period on grand-child's overweight/obesity and associated underlying pathways are uncertain. We examined whether the smoking status from the paternal line was associated with the grand-child's higher weight at birth, and overweight or obesity at 5 and 9 years of age. The grandparental smoking effect from the maternal line was also explored. Participants were fathers and grandparents and grand-children from the Lifeways Cross Generational Cohort (N = 1021 for the analysis at birth; N = 562 and N = 284 for the analysis at 5 and 9 years, respectively). Paternal and grandparental smoking was defined as smoking versus non-smoking. Children's weight categories compared were high versus normal weight at birth, and overweight/obesity versus normal weight (based on BMI and waist circumference) at age of five and nine years. Logistic regression models were used to estimate the crude and adjusted associations. After adjustment for several child and parental factors, at age five there was an association between paternal smoking and offspring's overweight/obesity based on BMI (Adjusted Odds Ratio (AOR), and 95%CI: 1.76, 1.14-2.71, p-value: 0.010), most marked for boys (AOR: 2.05, 1.06-3.96, p-value: 0.032). These associations remained when confined to the children sample with biological fathers only (overall sample, AOR: 1.92, 1.22-3.02, p-value: 0.005; son, AOR: 2.09, 1.06-4.11, p-value: 0.033). At age 9, the paternal grandmothers' smoking was positively associated with their grandchild's overweight/obesity status based on waist circumference (AOR: 3.29, 1.29-8.37), and especially with that of her granddaughter (AOR: 3.44, 1.11-10.69). These associations remained when analysing only the children sample with biological fathers (overall sample, AOR: 3.22,1.25-8.29, p-value: 0.016; granddaughter, AOR: 3.55, 1.13-11.15, p-value: 0.030). The smoking habit from the paternal line is associated with grand-children's adiposity measures during their early childhood, which might be epigenetically transmitted through male-germline cells.